In-vitro fragmentation of ovarian tissue activates primordial follicles through the Hippo pathway

What is the role of the Hippo and PI3K/Akt pathway in follicles during ovarian tissue culture in tissue derived from oncological patients and transgender men?

Results highlight a Hippo pathway driven primordial follicle activation in vitro, predominantly from Day 0 to Day 4.

In-vitro ovarian tissue culture aims at activating and maturing primordial follicles for fertility restoration in patients with a threatened ovarian reserve. Not all patients are eligible for ovarian cortex transplantation and therefore several groups are attempting to culture ovarian tissue in-vitro. Cortex fragmentation disrupts the Hippo pathway, leading to increased expression of downstream growth factors and follicle growth. The PI3K/Akt pathway is considered the intracellular pathway to where different extracellular factors involved in primordial follicle activation in-vivo converge. In order to optimise current ovarian tissue culture models, information on progression of these pathways during tissue culture is mandatory.

The first step of a multistep cortex culture system was performed using 144 ovarian cortex pieces from a total of six patients. Per patient, 24 cortical strips were cultured for 6 days and six pieces per patient were collected for downstream analysis of follicle development and Hippo and PI3K/Akt pathway targets every second day.

Ovarian tissue was obtained from oncological (N = 3; 28.67 ± 4.51 years) and transgender (N = 3; 23.33 ± 1.53 years) patients. Follicles were analysed using haematoxylin-eosin staining and pathways were studied using immunohistochemistry and precise follicle excision by laser capture micro-dissection for RT-qPCR analysis. MIQE guidelines for RT-qPCR were pursued. Reference gene selection (GAPDH, RPL3A, 18s rRNA) was performed using GeNorm Reference Gene Selection Kit. Statistical analysis was conducted with IBM SPSS Statistics 23 (Poisson regression, negative binomial regression, ANOVA and paired t-test).

Immunohistochemical analysis confirmed a Hippo pathway driven primordial follicle activation due to mechanical manipulation of the cortical strips. Ovarian tissue preparation and culture induced the inhibitory phosphorylated Yes-associated protein (pYAP) to disappear in granulosa cells of primordial follicles on Day 2. The stimulatory YAP on the contrary appeared in primordial granulosa cells over increasing culture days. Looking at the YAP target connective tissue growth factor (CTGF), a significantly up-regulated CTGF was noted in primordial follicles when comparing Day 2 and Day 4 (ratio Day 2/4 = 0.082; P < 0.05), clearly showing an effect on the Hippo pathway in primordial follicles during tissue culture. Follicle classification showed a significant drop in estimated primordial follicle counts in the oncological cohort (-78%; P = 0.021) on Day 2 and in the transgender cohort on Day 4 (-634%; P = 0.008). Intermediate follicle counts showed a non-significant increasing trend to during culture and this follicle recruitment and growth resulted in a significant rise in estimated primary follicle counts on Day 6 in oncological patients (170%; P = 0.025) and, although limited in absolute numbers, a significant increase in secondary follicles on Day 4 (367%; P = 0.021) in the transgender cohort. Subsequent antral follicle development could not be observed.

A limitation is the small sample size, inherent to this study subject, especially as a large amount of tissue was needed per patient to reduce inter-patient variation in different downstream analysis techniques. A particular and specific weakness of this study is the inability to include an age-matched control group.

These findings support an adapted tissue preparation for Hippo pathway disruption and a shorter first phase of tissue culture. This work may also have implications for transplantation of cryopreserved tissue as larger strips (and thus slower burnout due to less Hippo pathway disruption) could be a benefit.

This research was financially supported by the Foundation Against Cancer (Stichting tegen Kanker, TBMT001816N), the Flemish Foundation of Scientific Research (FWO Vlaanderen, FWO G0.065.11N10) and the Gender Identity Research and Education Society (GIRES) foundation. The authors declare no competing interests.

N/A.